Every voice is unique, like a fingerprint. So ETS uses the voice recordings from the Speaking section of the TOEFL iBT® test to help ensure that test takers aren’t misrepresenting themselves.
The recordings are compared using technology called voice biometrics, which uses statistical pattern-matching techniques, advanced voice classification methods, and inputs from multiple systems to compare speech samples from test takers.
Voice biometrics is used to identify potential testing misconduct in three main ways.
1. Repeat test takers – To compare voiceprints with speech samples from the same test taker across multiple test administrations. If the voiceprints are different, we investigate.
2. Potential impersonators within a test center – To compare voiceprints with speech samples for all test takers at a specific test center. If any voiceprints match, we investigate.
3. Known impersonators within a group – To compare voiceprints with speech samples from known imposters for all test takers within a suspect group. If any voiceprints match, we investigate.
Voice biometrics was introduced in 2012 for TOEFL® test security investigations, and is now used on all test takers worldwide before scores are released.
For more about TOEFL test security, visit https://www.ets.org/toefl/institutions/about/security/.